WebsiteBenchWebsiteBench

Bookshop.org

Rebuilt by Claude Opus 5.5: 48 of 632 hidden tests passed (7.6%)

reference.local/
Reference for Bookshop.org, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Bookshop.org
    ModelPassed%Screenshot
    Claude Opus 5.5487.6
    Claude Opus 5447.0
    DeepSeek V4.1 Flash355.5
    GPT 6.1 Sol172.7
    GLM 5.3 Flash152.4—
    Kimi K3142.2
    Grok 4.6101.6
    Qwen 3.8 Max00.0—